Published on Web 08/20/2010
Investigation of the Highly Active Manganese Superoxide Dismutase from Saccharomyces cerevisiae Kevin Barnese,†,‡,§ Yuewei Sheng,†,‡ Troy A. Stich,| Edith B. Gralla,‡ R. David Britt,| Diane E. Cabelli,*,⊥ and Joan Selverstone Valentine*,‡,§ Department of Chemistry and Biochemistry, UniVersity of California, Los Angeles, California 90095, Department of Bioinspired Chemistry, EWHA Womans UniVersity, Seoul 120-750, Korea, Department of Chemistry, UniVersity of California, DaVis, California 95616, and Department of Chemistry, BrookhaVen National Laboratory, Upton, New York 11973 Received May 14, 2010; E-mail:
[email protected];
[email protected] Abstract: Manganese superoxide dismutase (MnSOD) from different species differs in its efficiency in removing high concentrations of superoxide (O2-), due to different levels of product inhibition. Human MnSOD exhibits a substantially higher level of product inhibition than the MnSODs from bacteria. In order to investigate the mechanism of product inhibition and whether it is a feature common to eukaryotic MnSODs, we purified MnSOD from Saccharomyces cerevisiae (ScMnSOD). It was a tetramer with 0.6 equiv of Mn per monomer. The catalytic activity of ScMnSOD was investigated by pulse radiolysis and compared with human and two bacterial (Escherichia coli and Deinococcus radiodurans) MnSODs. To our surprise, ScMnSOD most efficiently facilitates removal of high concentrations of O2- among these MnSODs. The gating value k2/ k3 that characterizes the level of product inhibition scales as ScMnSOD > D. radiodurans MnSOD > E. coli MnSOD > human MnSOD. While most MnSODs rest as the oxidized form, ScMnSOD was isolated in the Mn2+ oxidation state as revealed by its optical and electron paramagnetic resonance spectra. This finding poses the possibility of elucidating the origin of product inhibition by comparing human MnSOD with ScMnSOD.
Manganese superoxide dismutase (MnSOD) enzymes catalyze superoxide (O2-) disproportionation by a mechanism that is more complex than those of the other SODs. In particular the reduction of superoxide can proceed via one of two pathways. One pathway dominates when the O2- concentration is low relative to the enzyme concentration (reaction 2), and the other pathway dominates when the ratio [O2-]/[MnSOD] is high (reactions 3 and 4). The paradoxical finding is that MnSOD is a less effective SOD catalyst when O2- levels are elevated.1 k1
2+ Mn3+SOD + O2 98 Mn SOD + O2 k2, 2H+
Mn2+SOD + O2- 98 Mn3+SOD + H2O2 k3
3+ 2Mn2+SOD + O2 98 Mn (OO )SOD k4, 2H+
Mn3+(OO2-)SOD 98 Mn3+SOD + H2O2
(1) (2)
(3) (4)
The depressed catalytic activity at high O2- concentrations is known to be due to the formation of a product-inhibited Mn3+†
These authors contributed equally. University of California, Los Angeles. Ewha Womans University, Korea. | University of California, Davis. ⊥ Brookhaven National Laboratory. ‡ §
10.1021/ja104179r 2010 American Chemical Society
peroxo adduct resulting from the inner sphere oxidation of Mn2+SOD by O2- (reaction 3). This pathway is considerably slower than the outer sphere oxidation and protonation pathway described in reaction 2. The relative levels of product inhibition are described kinetically by the value of k2/k3 for different MnSODs (Table 1). The contribution from this slower pathway is particularly pronounced in human MnSOD.1 Thus the O2- removal and H2O2 production rates are dependent on the relative levels of MnSOD and O2- and the degree of product inhibition of the specific MnSOD present. Table 1. Rate Constants for the Different MnSODs Organism
Human8b E. coli8c D. radiodurans8a S. cereVisiae (this work)
k1 (nM s-1)
k2 (nM s-1)
k3 (nM s-1)
k4 (s-1)
k2/k3
1.4 1.1 1.2 1.1-1.5
0.6 0.9 1.1 0.8
0.5 0.17 0.07 0.04-0.05
130 60 30 90-140
1.6 5.3 16 16-20
Superoxide concentrations are known to be variable in cells; for example, it has recently been shown that transient O2- bursts, termed “superoxide flashes”, are formed in human mitochondria,2 creating the possibility of even greater variability in H2O2 formation rates. However, a slower product-inhibited pathway in human MnSOD would allow for more constant H2O2 formation even when O2- concentrations vary. Low levels of H2O2 play an important role in signaling in mammalian cells, regulating numerous processes including rates of cell growth and division.3 It has been proposed that the slower pathway for human MnSOD appeared in response to an evolutionary pressure to control more tightly intracellular H2O2 levels,1 to reduce H2O2 mediated oxidative damage, and to optimize its signaling function.4 The k2/k3 values determined for human and bacterial MnSODs (Table 1) are consistent with this hypothesis. However, structural studies of different MnSODs, both wild type and mutant, have yet to reveal why k2/k3 differs so dramatically for this enzyme.1 The budding yeast Saccharomyces cereVisiae is widely used as a single-cell model for higher eukaryotic organisms because it is remarkably similar to mammalian cells. S. cereVisiae also appears to be less sensitive to H2O2 than human cells, and the only currently known H2O2 sensing proteins in S. cereVisiae (YAP1p and Skn7p) are involved in regulating oxidative stress protection; a more general signaling role has yet to be found.3,5 Both human MnSOD and S. cereVisiae MnSOD (ScMnSOD) are tetramers6 and localized to the mitochondrial matrix,7 while most bacterial MnSODs are dimers. J. AM. CHEM. SOC. 2010, 132, 12525–12527
9
12525
COMMUNICATIONS
Moreover, human MnSOD shares greater sequence similarity with ScMnSOD than with Escherichia coli or Deinococcus radiodurans MnSODs (62.2%, 52.9%, and 54.5% respectively).8 Published reports of the activity of ScMnSOD do not include a determination of the degree of product inhibition.6,9 We therefore turned our attention to characterizing the catalytic mechanism of ScMnSOD with the expectation that the high contribution from the productinhibited pathway would prove to be a property common to eukaryotic MnSODs. Surprisingly, we found instead that ScMnSOD is even less product-inhibited than the bacterial MnSODs characterized to date, surpassing even the high activity of MnSOD from the radiation resistant bacterium D. radiodurans.8a The gene of ScMnSOD,9 which includes the mitochondrial targeting sequence, was inserted into the plasmid YEp352. The enzyme was overexpressed in S. cereVisiae and purified using a protocol modified from that of Fridovich et al. (Supporting Information).6 The protein was isolated as a tetramer with the leader sequence removed and contained 0.6 equiv of Mn per monomer, as per size exclusion chromatography, mass spectrometry, and ICPMS, respectively. Although the fitting of the kinetic data reported here matches the experimental decay of pulsed O2-, out of the four rate constants, only k2 has been measured directly by pulse radiolysis. k2 was measured by oxidizing the resting enzyme with substoichiometric amounts of O2- and following the appearance of Mn3+SOD, which has a characteristic absorption band near 480 nm (Supporting Information). Unlike other MnSODs, ScMnSOD was isolated in the reduced state, precluding direct measurement of k1. k1, k3, and k4 were determined by fitting the observed rate of O2- loss (ε260 ) 2000 M-1 cm-1) at multiple enzyme concentrations (1-10 µM) and initial O2- concentrations (2-48 µM) using PRWIN, conditions that, while not physiological, are necessary to study kinetics by pulse radiolysis.1,8 The rate constants are compared to those known for other MnSODs in Table 1.
Figure 1. Decay of 2.3 µM (A) and 48 µM (B) O2- concentrations with MnSODs from different organisms (space permitting). ScMnSOD pulse radiolysis data are shown in black, while the lines are from Kintecus10 computer modeling using rate constants in Table 1: MnSOD from D. radiodurans (purple); E. coli (green); and human (blue); all at 1 µM Mn.
The ability of ScMnSOD to catalyze the dismutation of O2- was found to exceed that of the other characterized MnSODs. At low concentrations of O2- ([O2-] ≈ [MnSOD]), the observed rates of disappearance are similar for all the four enzymes (Figure 1A). However at high concentrations ([O2-] . [MnSOD]), the differences in activities are pronounced (Figure 1B). At the highest concentrations of O2- employed in pulse radiolysis, the order of the activities of the enzymes is S. cereVisiae > D. radiodurans > E. coli > human, which is identical to the ordering of k2/k3, indicating that the deciding factor is the tendency to form the product-inhibited state. Most known MnSODs rest in the Mn3+ oxidation state,11 but our ScMnSOD was consistently isolated as predominantly reduced Mn2+SOD; the electronic absorption spectrum of as-isolated Sc12526
J. AM. CHEM. SOC.
9
VOL. 132, NO. 36, 2010
MnSOD lacked the visible absorption band with a maximum near 480 nm. It was reported previously that the optical absorption spectrum of ScMn3+SOD was unusual relative to those of other MnSODs,6 but the spectrum of ScMn3+SOD that we obtained by reaction of as-isolated ScMnSOD with O2- using pulse radiolysis (Figure 2A, dots) was very similar to those of other Mn3+SODs, with an extinction coefficient of ∼800 M-1 cm-1. The small measured absorbance at 480 nm (Figure 2A, line) corresponds to greater than 90% of the as-isolated enzyme being in the Mn2+ state.
Figure 2. ScMnSOD was isolated in the Mn2+ oxidation state. (A) Electronic absorption spectra of as-isolated ScMnSOD (line) and Mn3+SOD obtained after reaction with O2- in pulse radiolysis (dots). (B) EPR spectrum of as-isolated ScMnSOD. Instrumental parameters: temperature, 4.7 K; microwave frequency, 9.685 GHz; microwave power, 0.2 mW; modulation amplitude, 0.8 mT.
The EPR spectrum of our as-isolated ScMnSOD also indicates that the enzyme is reduced, since the perpendicular-mode EPR spectrum is similar to those of other Mn2+SODs,12 with the usual six-line hyperfine splitting from the 55Mn nucleus (I ) 5/2) seen at geff ) 6.0 (Figure 2B). We looked for evidence of the integer spin Mn3+ (S ) 2) by parallel-mode EPR, but our spectrum lacked the sextet hyperfine pattern typically displayed by Mn3+SOD (Supporting Information).13 The only other MnSOD enzymes that have been isolated in the Mn2+ oxidation state are mutant MnSODs, most notably the Gln143 mutants of the human enzyme.14 However, the factors that determine the resting oxidation state are unknown. Increased levels of MnSOD activity have been shown to slow tumor growth in cultured human cells and in animal studies,15 and it has been proposed that this effect is related to cellular H2O2 levels. For that reason and to improve our understanding of the basis of the observed product inhibition, human MnSOD has been repeatedly mutated in attempts to make its activity resemble that of the bacterial proteins, but with limited success.1,8c As described above, human MnSOD already shares greater sequence similarity with ScMnSOD than with the bacterial enzymes, and the two eukaryotic proteins are tetramers while the bacterial ones are dimers. Also, ScMnSOD resembles the bacterial ones in that k3 is small but is similar to the human enzyme in that k4 is large. Thus, to improve our understanding of what causes the unusual kinetic properties of human MnSOD, it may be more productive to compare/contrast it with ScMnSOD than with the bacterial MnSODs. Investigation of the slight structural differences between the enzymes may provide a key to understanding the chemical mechanism of product inhibition. We will also continue to study the evolutionary significance of product inhibition by studying MnSOD from other organisms. Acknowledgment. This work was supported by Grant DK46828, KOSEF/MEST through WCU project (R31-2008-000-10010-0) to J.S.V. and National Institutes of Health and Grant GM48242 to R.D.B. Radiolysis studies were carried out at the Center for Radiation Chemistry Research at BNL, which is funded under Contract DE-AC02-98CH10886 with the U.S. Department of
COMMUNICATIONS
Energy and supported by its Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences. Supporting Information Available: ScMnSOD isolation details, pulse radiolysis, complete ref 2, and parallel mode EPR. This material is available free of charge via the Internet at http://pubs.acs.org.
(9) (10) (11) (12)
References (1) (2) (3) (4) (5) (6) (7)
Abreu, I. A.; Cabelli, D. E. Biochim. Biophys. Acta 2010, 1804, 263. Wang, W.; et al. Cell 2008, 134, 279. Veal, E. A.; Day, A. M.; Morgan, B. A. Mol. Cell 2007, 26, 1. Rhee, S. G. Science 2006, 312, 1882. Stone, J. R.; Yang, S. P. Antioxid. Redox Signaling 2006, 8, 243. Ravindranath, S. D.; Fridovich, I. J. Biol. Chem. 1975, 250, 6107. Luk, E.; Yang, M.; Jensen, L. T.; Bourbonnais, Y.; Culotta, V. C. J. Biol. Chem. 2005, 280, 22715. (8) (a) Abreu, I. A.; Hearn, A.; An, H.; Nick, H. S.; Silverman, D. N.; Cabelli, D. E. Biochemistry 2008, 47, 2350. (b) Hearn, A. S.; Stroupe,
(13) (14) (15)
M. E.; Cabelli, D. E.; Lepock, J. R.; Tainer, J. A.; Nick, H. S.; Silverman, D. N. Biochemistry 2001, 40, 12051. (c) Zheng, J.; Domsic, J. F.; Cabelli, D.; McKenna, R.; Silverman, D. N. Biochemistry 2007, 46, 14830. Schrank, I. S.; Sims, P. F. G.; Oliver, S. G. Gene 1988, 73, 121. Ianni, J. C. Kintecus, Windows ver. 3.95; 2008, http://www.kintecus.com/. Stroupe, M. E.; DiDonato, M.; Tainer, J. A. In Handbook of Metalloproteins; Messerschmidt, A., Ed.; John Wiley & Sons Ltd.: Chichester, U.K., 2001; Vol. 2, p 941. (a) Whittaker, J. W.; Whittaker, M. M. J. Am. Chem. Soc. 1991, 113, 5528. (b) Whittaker, M. M.; Whittaker, J. W. J. Biol. Chem. 1999, 274, 34751. Campbell, K. A.; Yikilmaz, E.; Grant, C. V.; Gregor, W.; Miller, A. F.; Britt, R. D. J. Am. Chem. Soc. 1999, 121, 4714. Leveque, V. J. P.; Stroupe, M. E.; Lepock, J. R.; Cabelli, D. E.; Tainer, J. A.; Nick, H. S.; Silverman, D. N. Biochemistry 2000, 39, 7131. (a) Church, S. L.; Grant, J. W.; Ridnour, L. A.; Oberley, L. W.; Swanson, P. E.; Meltzer, P. S.; Trent, J. M. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 3113. (b) Davis, C. A.; Hearn, A. S.; Fletcher, B.; Bickford, J.; Garcia, J. E.; Leveque, V.; Melendez, J. A.; Silverman, D. N.; Zucali, J.; Agarwal, A.; Nick, H. S. J. Biol. Chem. 2004, 279, 12769.
JA104179R
J. AM. CHEM. SOC.
9
VOL. 132, NO. 36, 2010 12527
